Comparison of endoscopic submucosal dissection and transanal endoscopic microsurgery for T1 rectal neuroendocrine tumors: a propensity score-matched study

Abstract

Endoscopic submucosal dissection (ESD) and transanal endoscopic microsurgery (TEM) are the most effective endoscopic resection methods for T1 rectal neuroendocrine tumors (NETs). We aimed to compare the efficacy and safety of ESD and TEM for rectal NETs≤20mm. Patients with rectal NETs≤20mm who underwent ESD or TEM were enrolled in this retrospective observational study. ESD and TEM groups were matched for pathologic tumor size and EMR history. We evaluated between-group differences in R0 resection rate, adverse event rate, procedure time, and hospital stay. We included 285 patients (ESD= 226, TEM= 59) in the final cohort, with 104 patients in the matched groups (ESD= 52, TEM= 52). The R0 resection rate was significantly higher for TEM (ESD 71.2% vs TEM 92.3%, P= .005). However, the median procedure time (ESD 22 [range, 11-65] vs TEM 35 [17-160] minutes, P< .001) and hospital stay (ESD 2.5 range 1-5] vs TEM 4 [3-8] days, P< .001) were significantly shorter for ESD. In the subgroup analysis of patients divided by tumor size<10mm (ESD= 218, TEM= 49) and 10 to 20mm (ESD= 8, TEM= 10)], there was no significant between-group difference in the R0 resection rate (83.5% vs 93.9%, P= .063 and 37.5% vs 80%, P= .145, respectively) or the rate of recurrence. Although TEM showed a better overall R0 resection rate for rectal NETs≤20mm, ESD could be a viable treatment modality concerning adverse events, procedure time, and hospital stay for rectal NETs<10mm with similar R0 resection rates in comparison with TEM.